PRIORITY CLAIM
This application claims priority from Italian Application for Patent No. TO2014A000503 filed Jun. 25 2014, the disclosure of which is incorporated by reference.
TECHNICAL FIELD
The present invention relates to a communication cell for an integrated circuit, to an electronic chip comprising said communication cell, to an electronic system including said chip, and to a test apparatus.
BACKGROUND
As is known, an integrated device or a generic electronic circuit in general comprises a plurality of input/output communication cells, which have the function of communication-interface means and are provided in the form of metal pads connected to a corresponding circuit configured to enable a bidirectional communication or, alternatively, a communication just for receiving data (input mode) or just for sending data (output mode) by/from the integrated device from/to the outside world. In the sequel of the description the term “communication cell” refers to the metal pad and to the communication circuit, whether unidirectional (input or output) or bidirectional (input and output), coupled to said metal pad.
Communication cells of a known type may be used, for example, for connecting components of a generic electronic circuit with elements external to the circuit itself (for example, for supplying to the integrated device control or data signals) also in a contactless (wireless) mode.
A communication cell of this type is provided, for example, in United States Patent Application Publication No. 2011/0171906 (incorporated by reference). However, said communication cell presents the disadvantage of being able to operate exclusively in contact mode or in capacitive contactless mode. Other contactless coupling modes are precluded. Furthermore, the operating mode must be defined beforehand and may not be defined automatically.
SUMMARY
An embodiment provides a communication cell for an integrated circuit, an electronic chip comprising said communication cell, an electronic system including said chip, and a test apparatus that will be able to extend the functionalities of the known art.
A communication cell for an integrated circuit, an electronic chip comprising said communication cell, an electronic system including said chip, and a test apparatus are provided.
In an embodiment, a communication cell for an integrated circuit comprises: a physical interface coupled to a first input terminal of the communication cell and configured to supply to the first input terminal an input signal, wherein, in a first operating condition, said input signal is a first operating signal supplied by a capacitive-signal source and, in a second operating condition, said input signal is a second operating signal supplied by an ohmic-signal source; and a first receiver circuit including a buffer configured to form an interface for reception of said second operating signal and generation of a second intermediate signal correlated to said second operating signal, a second receiver circuit including a transimpedance-amplifier circuit operatively coupled to said capacitive-signal source and configured to receive and amplify said first operating signal to generate a first intermediate signal; and an output stage configured to be coupled to said integrated circuit and including a selector device configured to receive the first intermediate signal and the second intermediate signal and to make available to said integrated circuit only the first intermediate signal in the first operating condition and only the second intermediate signal in the second operating condition.
In an embodiment, a communication method for an integrated circuit comprises the steps of: receiving, by a physical interface coupled to a first input terminal of the communication cell, an input signal, wherein, in a first operating condition, said input signal is a first operating signal supplied by a capacitive-signal source and, in a second operating condition, said input signal is a second operating signal supplied by an ohmic-signal source; and in the second operating condition, receiving by a buffer said second operating signal and generating a third intermediate signal correlated to said second operating signal, in the first operating condition, receiving and amplifying, by a second receiver circuit including a transimpedance-amplifier circuit, said first operating signal, thus generating a first intermediate signal; and receiving, by a selector device, the first intermediate signal and the second intermediate signal, and making available to said integrated circuit only the first intermediate signal in the first operating condition and only the second intermediate signal in the second operating condition.
BRIEF DESCRIPTION OF THE DRAWINGS
For a better understanding of the present invention a preferred embodiment thereof is now described, purely by way of non-limiting example, with reference to the attached drawings, wherein:
FIG. 1 shows a circuit diagram of a communication cell according to one aspect of the present invention;
FIG. 2 shows a circuit diagram of a receiver for digital signals detected in capacitive-coupling mode in the communication cell of FIG. 1;
FIGS. 3A-3D show, respectively, a signal transmitted from outside to the communication cell of FIG. 1, a signal at input to the receiver of FIG. 2, a signal at an intermediate step of processing by the receiver of FIG. 2, and a signal at output from the receiver of FIG. 2;
FIG. 4 shows a block diagram and a possible circuit embodiment of a receiver of digital signals that are detected in inductive-coupling mode in the communication cell of FIG. 1;
FIGS. 5A and 5B show a current signal that circulates on the transmitting antenna (FIG. 5B) induced on the basis of a digital voltage signal to be transmitted (FIG. 5A) generated by a transmitter external to the communication cell of FIG. 1;
FIGS. 5C and 5D show signals internal to the receiver of inductive signals of FIG. 4, during steps of processing of the signal of FIG. 5B;
FIG. 5E shows a signal at output from the receiver of inductive signals of FIG. 4, generated as a function of the signals of FIG. 5D;
FIG. 6 shows a transmission circuit designed to generate a signal that can be detected by the receiver of inductive signals of FIG. 4 and in particular to generate the current signal of FIG. 5B;
FIG. 7 shows an electronic circuit for detecting an operating mode of the communication cell of FIG. 1 chosen between ohmic-contact operating mode and capacitive contactless operating mode;
FIG. 8 shows a test device for an integrated circuit comprising a plurality of communication cells of FIG. 1;
FIG. 9 shows, in top plan view, a chip comprising a plurality of communication cells, which are arranged in a central portion of said chip;
FIG. 10 shows a plurality of chips, which comprise a respective plurality of communication cells of FIG. 2 connected together by said communication cells, thus enabling a 3D integration of said chips; and
FIG. 11 shows a communication cell according to an embodiment alternative to that of FIG. 1.
DETAILED DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic circuit representation of a communication cell 1 according to one aspect. The communication cell 1 is of an input/output (I/O) type—i.e., it may function both in input mode (in this case it receives a signal supplied thereto from outside)—and in output mode—i.e., it generates a signal. In greater detail, the communication cell 1 comprises a communication circuit 6 connected between an external-interface node 4, from which it receives an input signal Si, and an output port 11, designed to supply an output signal. The communication circuit 6 is operatively coupled to a plurality of physical-interface elements 9 via the external-interface node 4. The physical-interface elements 9 include, according to one embodiment, interface pads. The signal Si can be a data signal, or else a clock signal, or some other signal.
The communication circuit 6 includes a buffer circuit 10 (e.g., obtained with CMOS inverters cascaded to one another) and a selector device (multiplexer or mux) 24. In greater detail, the buffer circuit 10 is coupled between the external-interface node 4 and the selector device 24, whereas the latter is coupled between the output of the buffer circuit 10 and the output port 11 of the communication cell 1. The buffer 10 receives at input the signal Si and generates at output the signal Si′, which has matched voltage levels (for example, for enabling compatibility with the voltage levels accepted at input by the selector 24). Furthermore, the buffer 10 decouples the source of the signal Si from the integrated circuit 7, thus preventing the source of signal from absorbing current.
Via the output port 11 the selector device 24 interfaces with a generic electronic circuit 7.
The communication circuit 6 further comprises a driving circuit 2, which includes a buffer, configured to receive, via an input port 3 of the communication cell 1, a signal coming from the electronic circuit 7, and supply an output signal So for driving the external-interface node 4. In this way, the electronic circuit or integrated device 7 can communicate with the outside world, and the communication cell 1 thus operates in output mode.
The communication cell 1 is configured to enable two modes of external interfacing with the node 4: a direct-contact mode (ohmic contact) and a contactless or wireless mode. In particular, the contactless mode enables coupling of a capacitive type or, alternatively, of an inductive type at the external-interface node 4. For this purpose the physical interface 9 includes pads 9 a and 9 b and an antenna 5.
Capacitive coupling is obtained by operatively coupling the external-interface node 4 to a metal pad 9 a, which, in use, functions as plate of a capacitor (the other plate of the capacitor is provided by an external system for access to the pad). According to one embodiment, a dielectric layer completely covers the metal pad 9 a, to form the dielectric of the capacitor that, in use, comes to be formed between the external access system and the metal pad 9 a. According to a different embodiment, the dielectric layer is not present, and the dielectric between the plates of the capacitor that is formed in use is air.
Inductive coupling is obtained by providing the antenna 5 in the form of integrated coil connected to the node 4 preferably via a switch 5 b. The switch 5 b has the function of preventing (when open) setting-up of a resonant circuit formed by coupling of the antenna 5 with the capacitor formed, in use, by the pad 9 a and by possible parasitic capacitances present at the node 4. The switch 5 b is thus open when the system operates in capacitive mode, and closed when it is operating in inductive mode.
According to an embodiment provided by way of example, the antenna 5 is a quadrangular coil inductor provided on a number of metal levels, for example seven, to obtain a total inductance of some tens of nanohenries, the exact value of which depends upon the design specifications (for example, manufacturing process, communication distance, area occupied, etc.). It is evident that the antenna 5 can be obtained in different ways, in terms of geometry and metal levels; for example, the antenna 5 can be a planar antenna. Access from outside to said antenna 5 is obtained by inductive coupling with an external antenna belonging to an external access system.
Ohmic coupling is obtained by providing a metal pad 9 b designed to be ohmically contacted by an external system for access to the pad. This pad can be a further pad additional to the metal pad 9 a (for example, in the case where the latter is covered by a dielectric layer), or else can coincide therewith.
The metal pad 9 a that concurs in forming the capacitor for capacitive coupling is, according to one embodiment, directly coupled to the external-interface node 4. The antenna 5 for inductive coupling is coupled to the interface node 4 through the switch 5 b. Thus, in use, the input signal Si can be the capacitive signal generated by capacitive coupling, or else the inductive signal generated by inductive coupling, or else the signal generated by ohmic contact, according to the various possible operating conditions of the communication cell 1 of the present invention. The signal Si at input to the node 4 is a digital signal in the case where the communication is of an ohmic-contact type. Otherwise, if the communication is inductive or capacitive, the signal Si is of an analog type.
In order to enable operation in contactless mode, in particular for acquiring signals at input to the communication cell 1, the communication circuit 6 comprises a receiver device 22. The receiver device 22 includes: an input port 22 a connected with the external-interface node 4, from which it receives the input signal Si; a first control input 22 b, for receiving a control signal SW configured for identifying an operation of the receiver device 22 as capacitive or inductive receiver; and an output port 22 c, configured to supply an intermediate signal Sint at output from the receiver device 22. The intermediate signal Sint is a voltage signal of a digital type and is a function of the input signal Si.
The communication circuit 6 comprises a control input 22 d, which acts on the receiver device 22, for receiving an initialization signal SRESET configured to initialize the communication circuit 6 in one of the operating modes (for example, in a contactless operating mode).
The receiver device 22 comprises a first receiving circuit 26 and a second receiving circuit 28. The first receiving circuit 26 is configured to receive the input signal Si and processing it when the communication cell 1 operates in capacitive-coupling contactless mode, supplying at output a signal SCAP correlated to the signal received at input Si. The second receiving circuit 28 is configured to receive the input signal Si from the external-interface node 4 and processing it when the communication cell 1 operates in inductive-coupling contactless mode, supplying at output a signal SIND correlated to the signal received Si.
The control signal SW is generated, according to one embodiment, outside the communication cell 1 and is supplied to the communication cell 1 via an input port 23 of the communication cell 1 itself. The control signal SW is, for example, a signal having a first logic value designed to enable operation of just one between the first and second receiving circuits 26, 28, and a second logic value designed to enable operation of the other between the first and second receiving circuits 26, 28. The control signal SW can also be generated by a logic integrated in the communication cell 1. According to a further embodiment, the control signal SW is set in the design stage to a fixed value, according to whether it is desired to use the communication cell 1 in the inductive operating mode or, alternatively, in the capacitive operating mode (according to the particular application for which the communication cell 1 is designed).
The embodiment of FIG. 1 illustrates the presence of two coupling switches 27 a, 27 b, which are designed to couple/decouple the first and second receiving circuits 26, 28 to/from the selector device 24. According to a per se known embodiment, the switches 27 a and 27 b can be constituted each by pairs of n-MOS and p-MOS transistors.
The control signal SW is supplied to a respective control terminal of the coupling switches 27 a, 27 b for switching on just one of the coupling switches 27 a, 27 b and switching off the other of the coupling switches 27 a, 27 b. In this way, it is possible to select which between the signals SCAP and SIND is supplied at input to the selector device 24, thus supplying the intermediate signal Sint.
As an alternative or in addition to the presence of the coupling switches 27 a, 27 b, the control signal SW can have the function of switching off one between the first and second receiving circuits 26, 28, for example by uncoupling one between the first and second receiving circuits 26, 28 from the electrical supply (for instance, by driving a respective switch that couples/decouples the electrical supply to/from the first and second receiving circuits 26, 28).
With reference once again to the selector device 24, the latter includes: a first input 24 a connected with the external-interface node 4 via the buffer circuit 10, for receiving the signal Si′, which is correlated to the input signal Si; a second input 24 b, connected to the output port 22 c of the receiver device 22 for receiving the intermediate signal Sint (i.e., one of the signals SCAP and SIND); a control port 24 c, connected to an input port 29 of the communication cell 1, for receiving a selection signal SMUX designed to select which input 24 a, 24 b must be supplied to the output of the selector 24 (depending upon the type of operating mode, whether contact or contactless); and an output port 24 d, configured to supply at output from the selector device 24 alternatively the input signal Si received at the first input 24 a or the intermediate signal Sint, received at the second input 24 b (on the basis of the selection signal SMUX). The input signal Si, or intermediate signal Sint, at output from the selector device 24 is supplied to the output port 11 of the communication cell 1.
FIG. 2 shows, by a block diagram, a circuit designed to supply the first receiving circuit 26.
Present at the external-interface node 4, according to the representation of FIG. 2, is a coupling capacitor CC, which, in use, is formed between the external-interface node 4 of the communication cell 1 (for example, as has been said, formed by the metal pad 9 a) and a generic access device (provided with an own metal pad) used for accessing the communication cell 1 via capacitive coupling. This capacitor CC can have a value of capacitance of a few femtofarads (e.g., 3 fF for a distance between the plates that constitute the capacitor Cc of approximately 10 μm, with metal plates of a size of 50 μm×50 μm and air as dielectric). Further present between the external-interface node 4 and a ground reference terminal GND is a parasitic capacitor Cpar, having a value of capacitance variable on the basis of the manufacturing process, typically of a few picofarads (e.g., 1 pF).
The value of capacitance Cpar of the parasitic capacitor may not be determined precisely beforehand. Consequently, it is not possible to indicate precisely, beforehand, the amplitude of the input signal Si that is received on the interface node 4. In particular, said signal Si depends not only upon the signal transmitted externally, but also upon the capacitive divider H formed by the input coupling capacitor CC and by the parasitic capacitor Cpar. Said capacitive divider H is given by H=CC/(CC+Cpar).
The input signal Si is consequently attenuated by the value H. Since generally the value of capacitance Cpar is higher than the value of capacitance CC (for example, 150 times as much), the input signal Si is a signal with a swing that is excessively low to be received and interpreted without ambiguity.
The first receiving circuit 26 comprises a transimpedance amplifier 34, for example formed by an operational amplifier 32, having an inverting input 32 a coupled to the input-interface node 4 for receiving the input signal Si and a non-inverting input 32 b biased at a reference biasing voltage VBIAS. The reference biasing voltage VBIAS is chosen of a value such as to keep the operational amplifier in a high-gain regime.
Further present is a feedback branch 34 a including a resistor 36 that electrically couples the output terminal 32 c of the operational amplifier 32 to the inverting input 32 a via a switch 33 driven by a signal SMR. The signal SMR is obtained on the basis of the signals SMUX and SRESET in such a way that the signal SMR has a logic value “1” when SMUX has a logic value “0” and the reset signal is not active (SRESET=“0”), and SMR has a logic value “0” when SMUX=“1”, or following upon application of the signal SRESET. In the case provided by way of example considered herein we thus have SMR=(/SMUX AND /SRESET) where /SMUX is the signal SMUX negated, and /SRESET is the signal SRESET negated. When SMR=“1” the switch 33 is closed (i.e., it conducts current), whereas when SMR=“0” the switch 33 is open. In other words, during the capacitive-communication operating condition, the switch 33 is kept closed by the signal SMR=“1”, whereas in the remaining operating modes (i.e., inductive and ohmic-contact modes) the switch 33 is kept open by the signal SMR=“0”.
The signal SRESET drives the switch 33 into the open state when it is desired to operate in the inductive or ohmic-contact mode, thus preventing an “electrical conflict” on the input node 4.
The resistor 36 has an electrical resistance Rf of a value such as to amplify the analog input signal Si and not cause saturation of the operational amplifier and enable the correct operating swing; for example, with a two-stage circuit topology of the amplifier, the value is of the order of tens of kiloohms. The value of resistance Rf of the resistor 36 is chosen on the basis of the gain that it is desired to have on the input signal so that its output will have an amplitude such as to enable the circuits controlled by said output signal (e.g., the integrated circuit 7) to work properly.
It is pointed out that, in the capacitive operating mode, the input signal Si is the current that flows through the capacitance CC.
Furthermore, the first receiving circuit 26 comprises a buffer 38, for example provided by a hysteresis comparator, or a Schmitt trigger, which has the function of restoring the logic levels of the signal VX supplied at output from the transimpedance amplifier 34, generating at output a digital signal SCAP.
With reference to the transimpedance amplifier 34, during operation a feedback current if circulates through the resistor 36 that tends to bring the voltage value applied to the inverting terminal 32 a to the same voltage value VBIAS applied to the non-inverting terminal 32 b. The voltage signal VX available on the output of the operational amplifier 32 consequently has a plot similar to that of the current if, amplified by a value equal to Rf. The current if is equal to the current that circulates through the coupling capacitance CC minus the current that circulates through the parasitic capacitance Cpar. The latter is negligible because the operational amplifier 32 functions in a high-gain condition and the feedback keeps the voltage on the node 4 approximately constant. The current if is proportional to the derivative of a digital signal STX transmitted through the coupling capacitance CC.
It should be noted that the receiving circuit 26 thus provides a detection of the current generated by transition of the signal VTX (current sensing) that is independent of the value of the capacitive divider H mentioned previously.
FIGS. 3A-3D show by way of example the signal STX at input to the coupling capacitance CC (FIG. 3A), the input signal Si (in this case a current signal, FIG. 3B), the signal VX at output from the transimpedance amplifier 34 (FIG. 3C), and the signal SCAP correlated to the signal Si and with the rising and falling edges regenerated. For greater clarity, the signals of FIGS. 3A-3C are represented using the same time scale (axis of the abscissae).
With reference to FIG. 3B, it should be noted that the passage from the high logic value to the low logic value, and vice versa, of the input signal STX of FIG. 3 causes a corresponding variation of the signal Si, i.e., the current through the input terminal 4, which presents an impulsive plot.
FIG. 3D shows the signal SCAP generated at output from the buffer 38. The signal SCAP is, substantially, a delayed reconstruction of the input signal STX, and presents logic levels restored and a plot similar to that of the input signal STX of FIG. 3A. In this case, by way of example, the high logic value is approximately 2.5 V, and the low logic value is approximately 0 V, considering the values of the supply voltages used by the integrated circuit 7 of FIG. 1.
As already noted with reference to the receiver device 22 of FIG. 1, the latter further comprises the second receiving circuit 28, configured to receive input signals Si generated by inductive coupling. The external-interface node 4 is, in this case, coupled to the antenna 5 designed to acquire a signal generated inductively by an external transmission circuit (e.g., by supplying current pulses to a transmission antenna, or coil, using the circuit of FIG. 6).
FIG. 4 shows the second receiving circuit 28 according to one embodiment. The external-interface node 4 of FIG. 1 is, in the embodiment of FIG. 4, decomposed into the two nodes 4 a and 4 b. The node for ohmic or capacitive coupling is the node 4 a or 4 b indifferently.
Two switches 5 b′, 5 b″ provide the switch 5 b shown in FIG. 1 and enable decoupling of the antenna 5 from the communication circuit 6 of FIG. 1. The second receiving circuit 28 includes a first input and a second input 4 a, 4 b, which are designed to receive a differential voltage signal supplied by the receiving antenna 5, and an output 28 c, designed to supply a signal SIND that is a function of the differential signal received at the inputs 4 a, 4 b. The signal SIND is a digital signal, designed to be processed by the electronic circuit or integrated device 7. The receiving circuit 28 will be better described in what follows.
With reference to FIG. 6, there now follows a description of an example of transmission circuit 40 external to the second receiving circuit 28, configured to transmit the signal STX (digital signal) that is to be received and processed by the receiving circuit 28. The signal STX transmitted is of the type shown in FIG. 5A.
For this purpose, the transmission circuit 40 includes a first inverter 42 and a second inverter 44 provided according to a prior-art scheme by transistors of a P-channel and N-channel CMOS type (in FIG. 6 these are the devices M1, M2, M3, M4) supplied by the supply voltage Vd used in the transmission circuit. In a per se known manner, the first inverter 42 comprises a first transistor M1 and a second transistor M2, which are connected between a terminal biased at a voltage Vd (e.g., 2.5 V) and a reference terminal GND, for example at an approximately zero voltage. At the low logic value “0” of the signal STX, the transistor M1 is on, whereas the transistor M2 is off, and vice versa.
The second inverter 44 is similar to the first inverter 42 and comprises a respective first transistor M3 and a second transistor M4, which are connected between the terminal biased at a voltage Vd and the reference terminal GND. At the low logic value “0” of the signal STX the transistor M3 is on, whereas the transistor M4 is off, and vice versa.
Preferably, the inverters 42, 44 are controlled by a respective buffer 48, 49 having the function of receiving the signal STX and generate a control signal of the respective inverter 42, 44 correlated to the signal STX but having an appropriate swing for driving the transistors M1-M4 of the respective inverters 42, 44 into the respective on and off states, or vice versa.
The transmission circuit 40 further includes a delayer element 46, for example provided by a chain of inverters in even number, configured to introduce a temporal delay on the signal STX supplied to the inverter 44.
The output of the first inverter 42 is operatively coupled to a first conduction terminal of a transmitting antenna 50, for example provided in the form of planar coil, but it is evident, for the purposes of the present invention, that other embodiments are possible. The output of the second inverter 44 is operatively coupled to a second conduction terminal of the transmitting antenna 50.
Assume that, in an operating condition of the transmission circuit 40, the signal STX has a low logic value “0” or a high logic value “1”, the circuit 40 is in steady-state condition, and the two conduction terminals of the antenna 50 are both at a constant potential or value GND or value Vd (low logic value or high logic value). Thus, there is no electric current flowing (FIG. 5B) and the antenna 50 does not transmit any signal. At a rising edge of the signal STX (from the logic value “0” to the logic value “1”), the transistor M1 is off and the transistor M2 is on. The delayer circuit 46 enables the second inverter 44 to have the transistor M3 still on and the transistor M4 off.
Thus, during the interval defined by the delayer element 46, a current Iant flows through the transmitting antenna 50 (see the current pulse in FIG. 5B).
The opposite passage of logic state, i.e., from the high logic value “1” to the low logic value “0”, of the signal STX causes a respective flow of current through the transmitting antenna 50, with an opposite direction with respect to passage from the low logic value “0” to the high logic value “1” of the signal STX. FIG. 5B shows the generation of current pulses through the transmitting antenna 50 in both of the situations described.
The electromagnetic signal generated by the transmitting antenna 50 is received by the receiving antenna 5. The latter is obtained, as has been said previously, integrated on one or more metal levels, as a coil or with different geometrical shapes, having a first conduction terminal 5′ and a second conduction terminal 5″. As is known, when immersed in the electromagnetic field generated by the current that flows in the transmitting antenna 50, a differential voltage is induced on the receiving antenna 5, between the first and second conduction terminals 5′, 5″. There is thus generated a difference of potential between the nodes 4 a and 4 b. The condition just described arises when the switches 5 b′, 5 b″ are closed, i.e., when the communication circuit 6 of FIG. 1 functions in inductive mode. The first conduction terminal 5′ of the receiving antenna 5 is operatively coupled to the input 4 a of the second receiving circuit 28 through the switch 5 b′, whereas the second conduction terminal 5″ of the receiving antenna 5 is operatively coupled to the input 4 b of the second receiving circuit 28 through the switch 5 b″.
Coupled between the input 4 a and the input 4 b is a biasing circuit 57, configured to bias the nodes of the inputs 4 a, 4 b by a biasing voltage VBIAS. In FIG. 5C the value shown of VBIAS is approximately 1V, considering that values of 2.5 V and 0 V for the high and low supply voltages have been respectively chosen.
According to one embodiment, the biasing circuit 57 includes a first resistor 57 a and a second resistor 57 b between the input 4 a and the input 4 b. The electrical node 59 between the first and second resistors 57 a, 57 b is biased at the biasing voltage VBIAS, generated in a per se known manner by a voltage generator.
In this way, the value of voltage VP′, VP″ present on the respective inputs 4 a, 4 b is given by VBIAS plus a signal that is a function of the conduction current of the receiving antenna 5 (FIG. 5C).
The resistors 57 a, 57 b are, according to one embodiment, provided by p-MOS transistors driven into the on state, and biasing of the electrical node 59 is, for example, provided by a feedback inverter circuit coupled between the supply voltage and the reference voltage GND.
As may be noted from FIG. 5C, the signal VP′ present on the input 4 a shows a negative peak at the rising edge of the current signal supplied to the transmitting antenna 50 and a positive peak at the falling edge of the current signal supplied to the transmitting antenna 50. The signal VP″ present on the input 4 b has an opposite evolution. In fact, on the nodes of the inputs 4 a and 4 b there may be noted an evolution of the voltage that is proportional to the time derivative of the current that flows through the receiving antenna 5.
To return to FIG. 4, the receiving circuit 28 comprises a differential-amplifier stage 51 between the nodes 4 a and 4 b, including a first amplifier 600 and a second amplifier 620 (which are similar to one another in terms of circuit topology and functional parameters), and a hysteresis comparator 80.
A possible circuit embodiment of the amplifier 600 is illustrated in FIG. 4, and includes a first operational amplifier 60. The first operational amplifier 60 has the non-inverting input 60 a coupled to the input 4 a via a capacitor 64. This capacitor, having for example a value of capacitance of 50 fF, has the function of eliminating the d.c. component of the signal present on the input 4 a. In this way, the signal supplied to the non-inverting input of the first operational amplifier 60 is a signal variable in time deprived of its d.c. component. Furthermore, the non-inverting input of the first operational amplifier 60 is connected to a voltage signal VB, applied by a resistor 66. The voltage signal VB is chosen of a value such as to bias the operational amplifier in the high-gain area, whereas the value of resistance of the resistor 66 is chosen of a value such as to not jeopardize the dynamic performance of the circuit, for example, a value of 100 kΩ.
The inverting terminal 60 b of the first operational amplifier 60 is connected to the voltage signal VB via a capacitor 65, having for example a value of capacitance of 150 fF. In turn, the capacitor 65 is coupled in series to a resistor 67, which has, for example, a value of electrical resistance of 5 kΩ. The capacitor 65 has the function of decoupling the output of the operational amplifier 60 from the input voltage VB (i.e., rendering the value of the output signal independent of the value of VB). The output 60 c of the first operational amplifier 60 is fed back to the inverting terminal 60 b by a feedback branch that includes a resistor 68, having for example a value of electrical resistance of 45 kΩ, which concurs with the resistor 67 in defining a gain of the first operational amplifier, in a known way (gain=1+R2/R1).
In use, the ensemble including the capacitors 64, 65, the resistors 66, 67, 68, and the operational amplifier 60, to which the signals previously described are supplied, provides an amplification of the signal VP′ present on the input 4 a (signal SO1 of FIG. 5D).
The second amplifier 620 is, according to one embodiment, of the same type as the amplifier 600, and comprises components that are similar to those already described with reference to the amplifier 600 and that have the same function. Consequently, the second amplifier 620 will not be described in detail herein.
With reference to the second amplifier 620, it comprises: a second operational amplifier 62, similar to the first operational amplifier 60; a capacitor 75 and a resistor 77 (similar to the capacitor 65 and the resistor 67 described previously) connected together in series between the voltage terminal VB and the inverting input 62 b of the operational amplifier 62; a feedback branch coupled between the output 62 c and the inverting input 62 b of the operational amplifier 62, including a resistor 78 (similar to the resistor 68); a capacitor 74 (similar to the capacitor 64) electrically coupled between the input node 4 b and the non-inverting terminal of the operational amplifier 62; and a resistor 76 (similar to the resistor 66) coupled between the voltage VB and the non-inverting input 62 a.
The second operational amplifier 620 provides an amplification of the signal VP″ present on the input 4 b (signal SO2 of FIG. 5D).
The signals at output from the first and second amplifiers 600, 620 are designated in FIG. 4 by the references SO1 and SO2, respectively. These signals are further shown in FIG. 5D. The signals SO1 and SO2 are connected to respective inputs of a second comparator stage 80, in particular of a hysteretic differential type. The thresholds of the comparator are defined according to a prior-art scheme that takes into account the amplitudes of the signals SO1 and SO2. The comparator circuit 80 is, in particular, a differential sense amplifier with hysteresis. The transition of the output of the comparator circuit 80 is thus controlled by the difference signal SO1−SO2.
To return to FIG. 1, as described previously, the communication cell 1 can operate in contactless mode or else contact mode. As anticipated, one or other of the operating modes entails control of the selector device 24 by the signal SMUX. In the case of operation in contact mode, the signal SMUX enables the input 24 a of the selector device 24, transferring the input signal Si′ onto the output 24 d of the selector device 24; in the case of operation in contactless mode, the signal SMUX enables the input 24 b of the selector device 24, transferring the signal Sint supplied by the receiver device 22 onto the output 24 d of the selector device 24.
The signal SMUX is a digital signal and acts downstream of the switches 27 a and 27 b driven by the signal SW (FIG. 1), which discriminates between contactless mode of a capacitive type and contactless mode of an inductive type. In the contactless mode of a capacitive type, according to one aspect of the present invention, the signal SMUX is generated by a detection circuit 90 of the type shown in FIG. 7. According to one embodiment, the detection circuit 90 is formed in the integrated form in the communication cell 1 and thus forms part of the communication cell 1 itself.
According to a different embodiment, the detection circuit 90 can be external to the communication cell 1 and operatively coupled to the latter in such a way as to supply the signal SMUX.
The detection circuit 90 is configured to detect automatically whether an interface external to the communication cell 1, for example a probe or tip of a test apparatus, is in direct contact (ohmic contact) with the metal pad 9 b coupled to the external-interface node 4, or else, if it is coupled in contactless mode, to the metal pad 9 a.
The detection circuit 90 comprises an internal biasing block 92, having a high output resistance, which can be provided for example with a p-MOS transistor 92 a and an n-MOS transistor 92 b in below-threshold operating conditions according to a prior-art modality. The transistors 92 a, 92 b have their respective drain terminals connected together at an electrical node 93, whereas the source terminal of the transistor 92 a is coupled to VDD and the source terminal of the transistor 92 b is coupled to the ground reference voltage GND.
According to one embodiment of the present invention, the control signal SRESET initializes the communication circuit 1 in FIG. 1 in the contact operating mode and forces the switch 33 of FIG. 2 into the open state, interrupting the feedback branch 34 a. If, after the action of the initialization signal SRESET, the operating mode is capacitive, the external-interface node 4 remains floating. Consequently, the only biasing contribution of the input node 4 to the transistors 95 and 96 is the circuit 92, and the biasing value is fixed by the sizing of the transistors 92 a and 92 b chosen in such a way that the voltage on the conduction node 93, which coincides with the node 4, is of a value comprised between the high supply value VDD (e.g., 2.5 V) and the low reference value GND (e.g., 0 V) for keeping in conduction both the transistor 95 and the transistor 96.
The detection circuit 90 further comprises a current mirror 94 including a p-MOS transistor 94 a belonging to the input branch 97 of the current mirror 94, and a second p-MOS transistor 94 b belonging to the output branch 98 of the current mirror 94. The gate terminals of the first and second transistors 94 a, 94 b are electrically connected together and to the drain terminal of the transistor 94 a, thus providing the current mirror 94. The current that flows in the branch 97 of the current mirror 94 is thus fed back to the branch 98. The gate terminal of the transistors 94 a and 94 b is also connected to the gate terminal of the transistor 104, which is thus turned on if current circulates in the branch 97.
The first branch 97 of the current mirror 94 includes an inverter, which is formed by the p-MOS transistor 95 and an n-MOS transistor 96 and has the input connected to the terminal 93 that coincides with the terminal 4. In this way, in use, after application of the signal SRESET, if the operating mode is contactless, biasing of the gate terminals of the transistors 95 and 96 is guaranteed by the biasing voltage present on the conduction node 93 generated by the circuit 92, which is of a value, as previously described, such as to keep both of the transistors 95 and 96 in the on state.
A second current mirror 102 includes a transistor 102 a, coupled to the transistor 94 b in such a said that they have their own respective drain terminal in common. The source terminal of the transistor 102 a is biased at the low supply voltage, for example the ground voltage GND. Furthermore, the gate and drain terminals of the transistor 102 a are connected together. The second current mirror 102 further includes a transistor 102 b, having the gate terminal coupled to the gate terminal of the transistor 102 a. The source terminal of the transistor 102 b is biased at the ground reference voltage GND. Both of the transistors 102 a and 102 b are, in this embodiment, of an n-MOS type.
The drain terminal of the transistor 102 b of the second current mirror 102 is coupled to the drain terminal of the transistor 104, of a p-MOS type, having the gate terminal biased by the same biasing voltage as the transistors 94 a and 94 b of the current mirror 94.
The detection circuit 90 further comprises a comparator 106, having the inverting input terminal 106 a coupled to a reference-voltage generator (not shown) in order to generate a comparison voltage signal SCMP. The non-inverting terminal of the comparator 106 is instead coupled to an electrical node 107 defined between the transistors 94 b and 102 a, for receiving the voltage S107 present on the node 107. The comparison voltage signal SCMP has a value such as to discriminate between the voltage values present on the node 107 in the respective operating conditions in which current circulates or does not circulate in the branch 98.
The result of the comparison between the signals SCMP and S107 is indicative of the operating condition of the detection circuit 90. When the transistors 95 and 96 are both driven in conduction by the leakage current of the transistors 92 a, 92 b, the circuit operates in the condition where current circulates on the branch 97. This current is mirrored on the branch 98, and the voltage on the node 107 is brought to a value such that S107>SCMP and the output of the comparator 106 assumes a high logic value “1”, thus switching the transistor 112 off. The voltage value on the node 116 depends upon sizing of the transistors 104 and 102 b and is chosen in a way such that it is recognized as low logic value by the chain of inverters 120. The output of this chain of inverters 120 is the signal SMUX, which controls the selector 24 of FIG. 1, thus selecting the input Sint. Simultaneously, by controlling the switch 33 of FIG. 2, the signal SMUX closes the feedback 34 a of the transimpedance amplifier 32 (i.e., by driving the switch 33 by the signal SMR). Consequently, in this operating condition and once all the transients have terminated, the voltage on the node 4 (signal Si) is controlled at the value VBIAS. With reference once again to FIG. 7, the voltage VBIAS on the node 4 is such as to keep the transistors 95 and 96 both in conduction and keep the logic value of the signal SMUX at “0”.
Instead, if after application of the signal SRESET operation is according to the contact operating mode, the external-interface node 4 is forced by a transmitter circuit with low output impedance (for example, a digital buffer) and the signal Si assumes the high logic value “1” and the low logic value “0” of a digital signal, such as, for example, the signal STX of FIG. 3A. The value of said signal Si is such as to drive into the on state one of the transistors 95 and 96 and into the off state the other of the transistors 95 and 96, irrespective of the presence of the biasing signal generated by the transistors 92 a, 92 b. In other words, in steady-state conditions, the voltage values corresponding to the high and low logic values of the signal Si are such as to keep in conduction one and only one of the transistors 95 and 96. In this case, since one of the transistors 95 and 96 is always off, there does not flow current in the branch 97, and the current in the output branch 98 of the mirror 94 is zero and the transistor 104 is off.
The voltage value of the signal S107 is brought to a low value such as to force a zero current on the input of the current mirror 102, thus biasing the transistor 102 a in the off condition. In these conditions, the voltage value on the node 107 is low and guarantees S107≦SCMP. The output of the comparator 106 thus assumes the low logic value “0” sending into conduction the transistor 112, which is traversed by a current for charging the capacitor 110, thus sending the voltage value on the node 116 to the value VDD, and also the signal SMUX goes to the high value. With reference to FIG. 1, the selector 24 selects the signal Si′, and the feedback branch 34 a in FIG. 2 is open.
Switching-on and switching-off of the transistor 112 are thus controlled by the output of the comparator 106. In this way, when the output of the comparator is the low logic signal “0”, the transistor 112 is on and the capacitor 110 is charged; when the output of the comparator is the high logic signal “1”, the transistor 112 is off and the capacitor 110 is discharged.
The node 116 is connected to the output of the detection circuit 90. In other words, the output of the detection circuit 90 is the signal across the capacitor 110.
In order to guarantee a digital output signal having well-defined rising and falling edges, a chain of an even number of inverters 120 is present (here, two inverters 122, 123), configured to receive the signal on the node 116 and supply at output a digital signal correlated to the signal across the capacitor 110. The signal supplied at output from the chain of inverters is the signal SMUX described previously.
The presence of the capacitor 110 and of the transistors 104, 112 and 102 b guarantees that, during transition of the signal Si from the voltage values corresponding to the high and low logic values, and vice versa, the voltage on the node 116 may not drop to values such as to cause switching of the inverter 122 and, in particular, thus guarantees that the signal SMUX will remain constant at the high value. The capacitor 110 is thus sized in such a way that its discharge time is longer than the duration of the transient of the signal Si present on the input node 4.
Consequently, according to the embodiment described, a high logic value of SMUX identifies the contact operating mode; instead, a low logic value of SMUX identifies of the contactless operating mode, as indicated in FIG. 1.
The detection circuit 90 of FIG. 7 is also able to identify passage from a contact mode to a contactless mode. Following upon this passage, the voltage Si on the node 4 is no longer forced by an external low-impedance transmitter circuit, and thus its value is determined by the biasing circuit 92, exactly as in the case immediately subsequent to application of a reset as described previously.
The communication cell 1 described, in its totality or portions thereof, which thus guarantee one or more of the operating modes, finds various applications in electronic systems.
The communication cell 1 finds application in electronic systems of an integrated type that must be subjected to a testing step for evaluating a proper operation thereof. Electronic systems that comprise a plurality of communication cells 1 can interface with test probes indifferently of a contact or contactless type, according to the particular requirements and/or of the instrumentation available for carrying out the test.
FIG. 8 shows a test system 199 for a device under test 200 (for example, an integrated circuit). The test system 199 comprises an array of probes 201 (known as “probe card”) comprising a plurality of probes 202 configured to operate in contactless mode. In particular, the array of probes 201 comprises a plurality of probes 202 configured to capacitive and/or inductive couple with a respective plurality of communication cells 1 belonging to the device under test 200. Alternatively, the probes 202 can be configured to couple with a respective communication cell 1 both in contact mode and contactless mode. In this latter case, for example, each probe 202 comprises a terminal portion 202′ formed by a respective metal pad designed for capacitive coupling with the metal pad 9 a, or including an antenna designed to be positioned in the proximity of the antenna 5 of the communication cell 1. In the latter case, each probe 202 further comprises a circuit 40 of the type shown in FIG. 6, in order to be able to communicate in data transmission, with inductive coupling, with a communication cell 1 of the device under test 200.
Each probe 202 is further connected to a test station 206, from which it receives test signals for the device under test 200 and to which it sends signals of response of the device under test 200 to the test signals so that they can be analyzed.
In addition, as shown in FIG. 9, testing communication cells 1 can be arranged in any area of a generic chip 301. For example, the chip 301 comprises a central portion 401 surrounded by a peripheral portion 402. In particular, one or more of the communication cells 1 can be arranged within the central portion 401, and be in any case easily accessible by capacitive or inductive connection. This is particularly advantageous in integrated chips or circuits of a pad-limited type, in which the number of communication cells 1 that can be obtained in the peripheral portion 402 of the chip 301 is limited.
Further, an additional application regards assembly of one or more chips in a package according to a mode with physical contact obtained by microbumps (FIG. 10). In this mode of use, the communication cell 1 previously described (provided with the detection circuit 90 of FIG. 7) can be in the contact operating mode or, indifferently, in the contactless operating mode of a capacitive type. The pads 9 a and 9 b coincide with one another in a single common metal pad. The microbump process, according to the known art, is carried out at this common pad 9 a/9 b.
FIG. 10, shows, according to the present invention, a first chip 601 connected to a second chip 602 via a plurality of electrical-connection elements 604 (for example, bumps, stubs, through-silicon vias, etc.). The second chip 602 is thus connected to a substrate 606 via a respective plurality of electrical-connection elements 605. Furthermore, each electrical- connection element 604, 605 is interfaced with the respective chip 601, 602 and with the substrate 606 via a communication cell 1 of the type described previously, which is provided with the circuit 90 of FIG. 7, as has been said. In this way, if, for example on account of the variations of manufacturing process or oxidation, one or more electrical- connection elements 604, 605 do not have an adequate thickness to enable an ohmic connection with the respective electrical- connection element 604, 605 that it faces (see, for example, in FIG. 10, the electrical- connection elements 604, 605 on the right), the communication in transmission and reception is in any case guaranteed in capacitive-coupling contactless mode according to what has already been described previously. In this way, it is evident that the reliability and the production yield are considerably increased.
From an examination of the characteristics of the invention provided according to the present disclosure, the advantages that it affords are evident.
In the capacitive transmission mode, the receiving circuit 26 provides a detection of the current generated by the transition of the signal VTX (current sensing) and not of the voltage attenuated by the capacitive divider H, as occurred instead in the situation described in US 2011/0171906. This renders the sensitivity of the circuit proposed much more independent of the value of the parasitic capacitances present on the input node 4 and thus enables correct reception for lower values of the coupling capacitance CC.
In particular, the detection circuit 90 of FIG. 7 enables detection of the ohmic-contact operating mode or of the capacitive contactless mode and consequently correct forcing of the selection signal of the selector device 24.
The communication cell 1 can be used instead of standard digital library I/O pads for enabling electrical wafer sorting (EWS) of electronic devices according to the modality that, according to the circuit to be tested, is more convenient, thus obtaining an increase of the yield of the testing process.
Furthermore, the communication cell 1, for example deprived of the antenna 5 and of the circuit 28, can be used instead of standard library I/O pads for enabling proper communication between the integrated device 7 and other contact-mode devices, for example via microbumps, also in the presence of non-idealities of the process of creation of the ohmic interconnections themselves, thus obtaining an increase of the yield of the packaging process. In fact, in the case of absence of ohmic contact, the communication is in any case possible by exploiting in an automatic way the capacitive communication mode.
Pads enable access to generic integrated circuits without the aid of probes of a mechanical type, which operate in direct electrical contact. Furthermore, they enable intra-chip communication and 3D integration in an economically advantageous way of compact chips without electrical wire connections. Furthermore, the transfer of data between communication cells according to the invention is possible also at high frequency, rendering possible communication, for example, between different chips at a high speed.
Finally, it is clear that modifications and variations may be made to the invention described and illustrated herein, without thereby departing from the scope thereof, as defined in the annexed claims.
According to one embodiment, shown in FIG. 11, the communication cell 1 (in FIG. 11 designated by the reference 1′) comprises just the conductive pads 9 a and 9 b (possibly coinciding with one another in a single pad), and not the antenna 5. In this case, the detection circuit 28 is not present either, and the communication cell 1 operates only in the ohmic-contact and capacitive mode. The remaining circuit elements of the communication cell 1′ of FIG. 11 correspond to those already illustrated with reference to the communication cell 1 of FIG. 1 and are not described any further herein.
The communication cell 1′ integrates, according to one embodiment, the detection circuit 90 of FIG. 7. Furthermore, the communication cell 1′ can be used instead of the communication cell 1 in the embodiments described with reference to FIGS. 8, 9, and 10.
According to a different embodiment (not shown in the figures), the communication cell of the present invention comprises the conductive pad 9 b and the antenna 5, and not the pad 9 a. In this case, the detection circuit 26 is not present either, and the communication cell 1 operates in just the ohmic-contact and inductive mode. In this embodiment, the circuit 90 of FIG. 7 can be omitted.
According to a further embodiment (not shown in the figures), alternative to the previous one, the communication cell of the present invention comprises the conductive pad 9 a and the antenna 5, and not the pad 9 b. In this case, the buffer 10 can be omitted, and the communication cell operates just in the capacitive and inductive modes.